abkit 0.1.0__tar.gz → 0.1.1__tar.gz

{abkit-0.1.0/src/abkit.egg-info → abkit-0.1.1}/PKG-INFO

@@ -1,7 +1,7 @@
 Metadata-Version: 2.4
 Name: abkit
-Version: 0.1.0
-Summary: Audit an A/B-test readout before you ship the decision
+Version: 0.1.1
+Summary: Sample-ratio, peeking, multiple-testing and winner's-curse checks for experiment readouts
 Author: Hadi Mohammadi
 License: MIT
 Project-URL: Homepage, https://github.com/mohammadi-hadi/abkit
@@ -30,9 +30,12 @@ Dynamic: license-file
 
 # abkit
 
-Audit an A/B-test readout before you ship the decision.
+Sample-ratio, peeking, multiple-testing and winner's-curse checks for experiment readouts.
 
 [![ci](https://github.com/mohammadi-hadi/abkit/actions/workflows/ci.yml/badge.svg)](https://github.com/mohammadi-hadi/abkit/actions/workflows/ci.yml)
+[![PyPI](https://img.shields.io/pypi/v/abkit)](https://pypi.org/project/abkit/)
+[![DOI](https://zenodo.org/badge/DOI/10.5281/zenodo.21809478.svg)](https://doi.org/10.5281/zenodo.21809478)
+[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](LICENSE)
 
 An experiment readout can say "B wins, p = 0.03, ship it" while the traffic
 split was broken, three whale customers carried the whole effect, the analyst
@@ -96,7 +99,7 @@ the exploratory split of the [Upworthy Research Archive](https://doi.org/10.1038
 ## Install
 
 ```
-pip install git+https://github.com/mohammadi-hadi/abkit
+pip install abkit
 ```
 
 Python 3.11+. Runtime dependencies: numpy, pydantic, matplotlib.
@@ -239,19 +242,25 @@ uncertainty on every number and validation by implantation.
 
 Companion projects: [judgekit](https://github.com/mohammadi-hadi/judgekit)
 applies the same audit-before-you-trust pattern to LLM judges;
-[trajectory-judge](https://github.com/mohammadi-hadi/trajectory-judge)
-measures what outcome-only judges miss on agent trajectories.
+[raterkit](https://github.com/mohammadi-hadi/raterkit) applies it to the human
+labels underneath both; [trajectory-judge](https://github.com/mohammadi-hadi/trajectory-judge)
+measures what outcome-only judges miss on agent trajectories;
+[arenakit](https://github.com/mohammadi-hadi/arenakit) applies it to pairwise
+model leaderboards.
 
 ## Citation
 
-If abkit is useful in your work, please cite it (see
-[CITATION.cff](CITATION.cff)):
+Releases are archived on Zenodo. Cite the concept DOI
+[10.5281/zenodo.21809478](https://doi.org/10.5281/zenodo.21809478), which always
+resolves to the latest version; structured metadata is in
+[CITATION.cff](CITATION.cff).
 
 ```bibtex
 @software{mohammadi_abkit,
   author  = {Mohammadi, Hadi},
   title   = {abkit: audit an A/B-test readout before you ship the decision},
   url     = {https://github.com/mohammadi-hadi/abkit},
+  doi     = {10.5281/zenodo.21809478},
   version = {0.1.0},
   year    = {2026}
 }

{abkit-0.1.0 → abkit-0.1.1}/README.md

@@ -1,8 +1,11 @@
 # abkit
 
-Audit an A/B-test readout before you ship the decision.
+Sample-ratio, peeking, multiple-testing and winner's-curse checks for experiment readouts.
 
 [![ci](https://github.com/mohammadi-hadi/abkit/actions/workflows/ci.yml/badge.svg)](https://github.com/mohammadi-hadi/abkit/actions/workflows/ci.yml)
+[![PyPI](https://img.shields.io/pypi/v/abkit)](https://pypi.org/project/abkit/)
+[![DOI](https://zenodo.org/badge/DOI/10.5281/zenodo.21809478.svg)](https://doi.org/10.5281/zenodo.21809478)
+[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](LICENSE)
 
 An experiment readout can say "B wins, p = 0.03, ship it" while the traffic
 split was broken, three whale customers carried the whole effect, the analyst
@@ -66,7 +69,7 @@ the exploratory split of the [Upworthy Research Archive](https://doi.org/10.1038
 ## Install
 
 ```
-pip install git+https://github.com/mohammadi-hadi/abkit
+pip install abkit
 ```
 
 Python 3.11+. Runtime dependencies: numpy, pydantic, matplotlib.
@@ -209,19 +212,25 @@ uncertainty on every number and validation by implantation.
 
 Companion projects: [judgekit](https://github.com/mohammadi-hadi/judgekit)
 applies the same audit-before-you-trust pattern to LLM judges;
-[trajectory-judge](https://github.com/mohammadi-hadi/trajectory-judge)
-measures what outcome-only judges miss on agent trajectories.
+[raterkit](https://github.com/mohammadi-hadi/raterkit) applies it to the human
+labels underneath both; [trajectory-judge](https://github.com/mohammadi-hadi/trajectory-judge)
+measures what outcome-only judges miss on agent trajectories;
+[arenakit](https://github.com/mohammadi-hadi/arenakit) applies it to pairwise
+model leaderboards.
 
 ## Citation
 
-If abkit is useful in your work, please cite it (see
-[CITATION.cff](CITATION.cff)):
+Releases are archived on Zenodo. Cite the concept DOI
+[10.5281/zenodo.21809478](https://doi.org/10.5281/zenodo.21809478), which always
+resolves to the latest version; structured metadata is in
+[CITATION.cff](CITATION.cff).
 
 ```bibtex
 @software{mohammadi_abkit,
   author  = {Mohammadi, Hadi},
   title   = {abkit: audit an A/B-test readout before you ship the decision},
   url     = {https://github.com/mohammadi-hadi/abkit},
+  doi     = {10.5281/zenodo.21809478},
   version = {0.1.0},
   year    = {2026}
 }

{abkit-0.1.0 → abkit-0.1.1}/pyproject.toml

@@ -4,8 +4,8 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "abkit"
-version = "0.1.0"
-description = "Audit an A/B-test readout before you ship the decision"
+version = "0.1.1"
+description = "Sample-ratio, peeking, multiple-testing and winner's-curse checks for experiment readouts"
 readme = "README.md"
 license = {text = "MIT"}
 authors = [{name = "Hadi Mohammadi"}]

{abkit-0.1.0 → abkit-0.1.1}/src/abkit/__init__.py

@@ -1,11 +1,11 @@
-"""abkit: audit an A/B-test readout before you ship the decision."""
+"""abkit: sample-ratio, peeking, multiple-testing and winner's-curse checks."""
 
 from .audit import run_audit
 from .io import dump_design, dump_units, load_design, load_units
 from .result import AuditResult, ProbeResult, SkippedProbe
 from .schema import Design, Look, Unit
 
-__version__ = "0.1.0"
+__version__ = "0.1.1"
 
 __all__ = [
     "AuditResult",

{abkit-0.1.0 → abkit-0.1.1}/src/abkit/cli.py

@@ -14,7 +14,9 @@ from .report import inject_readme, write_report
 
 def main(argv: list[str] | None = None) -> int:
     parser = argparse.ArgumentParser(
-        prog="abkit", description="Audit an A/B-test readout before you ship the decision."
+        prog="abkit",
+        description="Sample-ratio, peeking, multiple-testing and winner's-curse checks "
+        "for experiment readouts.",
     )
     parser.add_argument("--version", action="version", version=f"abkit {__version__}")
     sub = parser.add_subparsers(dest="command", required=True)

{abkit-0.1.0 → abkit-0.1.1/src/abkit.egg-info}/PKG-INFO

@@ -1,7 +1,7 @@
 Metadata-Version: 2.4
 Name: abkit
-Version: 0.1.0
-Summary: Audit an A/B-test readout before you ship the decision
+Version: 0.1.1
+Summary: Sample-ratio, peeking, multiple-testing and winner's-curse checks for experiment readouts
 Author: Hadi Mohammadi
 License: MIT
 Project-URL: Homepage, https://github.com/mohammadi-hadi/abkit
@@ -30,9 +30,12 @@ Dynamic: license-file
 
 # abkit
 
-Audit an A/B-test readout before you ship the decision.
+Sample-ratio, peeking, multiple-testing and winner's-curse checks for experiment readouts.
 
 [![ci](https://github.com/mohammadi-hadi/abkit/actions/workflows/ci.yml/badge.svg)](https://github.com/mohammadi-hadi/abkit/actions/workflows/ci.yml)
+[![PyPI](https://img.shields.io/pypi/v/abkit)](https://pypi.org/project/abkit/)
+[![DOI](https://zenodo.org/badge/DOI/10.5281/zenodo.21809478.svg)](https://doi.org/10.5281/zenodo.21809478)
+[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](LICENSE)
 
 An experiment readout can say "B wins, p = 0.03, ship it" while the traffic
 split was broken, three whale customers carried the whole effect, the analyst
@@ -96,7 +99,7 @@ the exploratory split of the [Upworthy Research Archive](https://doi.org/10.1038
 ## Install
 
 ```
-pip install git+https://github.com/mohammadi-hadi/abkit
+pip install abkit
 ```
 
 Python 3.11+. Runtime dependencies: numpy, pydantic, matplotlib.
@@ -239,19 +242,25 @@ uncertainty on every number and validation by implantation.
 
 Companion projects: [judgekit](https://github.com/mohammadi-hadi/judgekit)
 applies the same audit-before-you-trust pattern to LLM judges;
-[trajectory-judge](https://github.com/mohammadi-hadi/trajectory-judge)
-measures what outcome-only judges miss on agent trajectories.
+[raterkit](https://github.com/mohammadi-hadi/raterkit) applies it to the human
+labels underneath both; [trajectory-judge](https://github.com/mohammadi-hadi/trajectory-judge)
+measures what outcome-only judges miss on agent trajectories;
+[arenakit](https://github.com/mohammadi-hadi/arenakit) applies it to pairwise
+model leaderboards.
 
 ## Citation
 
-If abkit is useful in your work, please cite it (see
-[CITATION.cff](CITATION.cff)):
+Releases are archived on Zenodo. Cite the concept DOI
+[10.5281/zenodo.21809478](https://doi.org/10.5281/zenodo.21809478), which always
+resolves to the latest version; structured metadata is in
+[CITATION.cff](CITATION.cff).
 
 ```bibtex
 @software{mohammadi_abkit,
   author  = {Mohammadi, Hadi},
   title   = {abkit: audit an A/B-test readout before you ship the decision},
   url     = {https://github.com/mohammadi-hadi/abkit},
+  doi     = {10.5281/zenodo.21809478},
   version = {0.1.0},
   year    = {2026}
 }